Fig. 9.1
(a) An ileal conduit should have a protruding stoma and be sited through the rectus sheath. (b) The stoma should not be sited in a crease line or belt line. (c) The stoma protrudes nicely above the ring of the appliance, allowing for improved drainage into the stomal bag. (d) The wafer should be cut such that there is no pressure on the delicate mucosa while ensuring that minimal peristomal skin is exposed
Perioperatively the main considerations with ileal conduit stoma are bleeding and the risk of bowel necrosis. Massive bleeding is rare but if it occurs is usually associated with stomal varices that form through branches of the superior mesenteric vein and usually form within 2–3 years of surgery. While it is possible for these to form as a result of aberrant scarring, these are typically associated with portal hypertension secondary to hepatic dysfunction. In these patients this bleeding is typically refractory to local control with compression or suture ligation, and even “mucocutaneous disconnection” will fail in the long run due to the inevitable recanalization that will occur with ongoing high portal pressures [12]. In these cases transjugular intrahepatic portosystemic shunt (TIPS) has been successfully performed [13]. However, TIPS carries a 30% risk of hepatic encephalopathy and a high mortality rate in Child-Pugh B and C patients. In these cases, embolization and angiographic sclerosis have been described [14].
Ischemia and even necrosis of the bowel segment used is another serious complication that can be encountered. It is crucial during the harvesting of the distal ileal segment that the blood supply is identified and carefully maintained. It is our practice to take the distal end at the avascular line of Treves, thereby preserving an arcade to what will be the distal end of the bowel anastomosis as well as to the stomal end of the conduit. Ischemia can occur as a result of strangulation of the bowel at the level of the fascia, so it is important to pay attention to this. The stoma will be edematous and may appear dusky due to congestion for the first 2–3 days, but the mucosa should still blanch or pink up with touch and should become progressively pinker. Necrosis of the mucosa can also occur due to poor stomal care such as an appliance that compresses the mucosa instead of the surrounding skin only [15]. Patients who have undergone pelvic irradiation are especially susceptible to this, as the distal ileum may have experienced some treatment effects. It should be noted that while several relatively small case series have noted higher complication rates in IC patients with prior radiation, the actual stomal complications are negligible, at only two patients in one study. All of these were managed conservatively and not associated with full-thickness necrosis below the level of the fascia [16]. A transparent appliance is always recommended for ease of examination of the stoma. When mucosal level necrosis does occur, the superficial tissues will slough within a few weeks of improved stomal care. However, there is a risk of stomal retraction or stenosis in the long term. If full-thickness necrosis is suspected, this can be investigated at the bedside by gentle placement of a test tube into the stoma and illumination to assess the mucosa below the fascia. If this is indeterminate, flexible cystoscopy is an excellent way to visualize the entire luminal surface of the conduit. In the extremely rare situation that true bowel necrosis has occurred, emergent laparotomy must be pursued.
Long-term complications of the ileal conduit stoma include parastomal hernia, stomal stenosis, and retraction. The impact of the type of stoma constructed on these complications is debated. Historically, the Turnbull loop stoma has been reported to have a lower incidence of stomal stenosis but a higher incidence of parastomal hernia [17]. However, several large series comparing loop and end stomas specifically within ileal conduits found comparable rates of parastomal hernia in end and loop stomas [18–20].
Robotic-assisted radical cystectomy (RARC) is becoming more commonplace, although only 18% of diversions are being performed completely intracorporally [21]. In the case of ileal conduit, the bowel segment is harvested and prepared robotically, and typically the robot is undocked, and then the conduit is brought up and the stoma created in the typical fashion. Necrosis of the conduit due to inappropriate orientation has been cited in the literature; establishing and maintaining orientation during the robotic portion is key since the conduit itself is not examinable during stomal construction.
The International Robotic Cystectomy Consortium (IRCC) offers the largest series to date of RARC, comprised of over 900 patients from tertiary and community centers undergoing RARC from 2003 to 2011; 72% of these were ileal conduit. A 2014 analysis of intracorporeal versus extracorporeal diversion outcomes in this group by Ahmed et al. found lower complication, readmission, and transfusion rates for the intracorporeal group, although they had a higher 30-day reoperation rate at 8% [22]. The overall 30-day reoperation rate was 7%; cause of reoperation was not included in the database. While the intracorporeal group complications were lower than the extracorporeal group, it may be explained by the fact that more experienced robotic surgeons are more likely to perform the former. In addition, the 90-day mortality rates of both groups (1.6% and 4.9%, respectively) are higher than seen in other RARC or open RC series [21]. The IRCC data suggests that in the hands of experienced robotic surgeons, intracorporeal diversion has reasonable outcomes.
Differences between open and robotic conduit stomal rates of hernia or stenosis have not been reported, but are unlikely to be different since stomal construction is performed by open technique in both.
Parastomal Hernia in the Incontinent Stoma
Parastomal hernia is defined as a fascial defect allowing for protrusion of peritoneum or bowel segment. Although it may be diagnosed clinically with Valsalva maneuver and digital exam, more often the diagnosis is made using cross-axial imaging. The majority of patients are asymptomatic but occasionally may present with parastomal pain, symptoms of intermittent obstruction and, rarely, of bowel strangulation.
Widely accepted risk factors for development of a parastomal hernia are the same as those for development of an incisional hernia: obesity, poor nutrition, prolonged ileus, steroid use, wound infection, previous surgery, or radiation. However, this is largely anecdotal, and the published data is mixed. In a 2007 paper examining 137 radical cystectomy ileal conduit patients, an overall stomal complication rate of 15.7% was found of which the most common complication was parastomal hernia (91% of complications and 14% incidence overall); one went on to develop stomal prolapse [17]. The only predictor found was BMI, where obese patients had a complication rate of 27.3% compared to 4.1% in those with a normal BMI. Gender, age, and smoking status, prior radiation, did not predict stomal complications or parastomal hernia. However, this was just a univariable analysis. Similar outcomes were found in another series of 97 patients where BMI, diabetes, and emergent surgery predicted complications such as poor siting and skin excoriation, but no hernia or stenosis [23]. One third of parastomal hernia patients in this series went on to surgical repair for symptoms, of which 50% had a recurrence. This is similar to other reported rates of recurrence and highlights the difficulty in management of those patients who are symptomatic.
Given this, prevention of parastomal hernia is very important. Placement of the stoma through the rectus muscle and not lateral to it showed a reduction in incidence of hernia from 22% to 2% in a study of 130 patients [24], although other series did not find a difference [25, 26]. Nevertheless, given the absence of any randomized studies, we feel that surgeons should adhere to the principle of placement through the rectus muscle and rectus sheath. In addition, it is important not to make the fascial opening too large. Our practice is to create a defect that just barely allows two fingers for a loop conduit, which is just enough to accommodate the typical ileal diameter snugly. Similarly, the posterior fascial opening should be opened only enough to accommodate the bowel snugly, as enlargement of this has also been postulated to contribute to hernia formation.
A study of 782 ostomy patients from a large French database in 2011 found a parastomal hernia rate of 25.6% in urostomies (loop type, n = 180), which was lower than that seen in colostomy (28%) and higher that seen in ileostomy (16%), with a median time to diagnosis of 18 months.
Overall, the incidence of parastomal hernia in ileal conduit varies widely, with some reporting rates as low as 4.5 [27]; the variability in how it is defined, and the fact that two thirds of patients are asymptomatic and so may never present, may explain this. In a 1975 review, Leadbetter points out that while its incidence in the ileal conduit may be less than in colostomy, its propensity to cause symptoms due to urine leakage and resultant difficulty with appliance placement and skin irritation is higher [28]. While, as is true for ileostomy patients, the vast majority of ileal conduit patients with a parastomal hernia are asymptomatic, this may explain a higher rate of reoperation of around 30% compared to 10–20% in large series of all parastomal hernias [17, 29]. Additionally, the presence of a hernia adds to the psychological impact of a new stoma on the patient and has been associated with lower QOL after cystectomy [7].
In the largest series to date of parastomal hernia in ileal conduit after radical cystectomy patients specifically, Donahue et al. reviewed records of 433 patients who underwent surgery between 2006 and 2010. In order to establish a true incidence rate, all CT scans performed for oncologic follow-up were reviewed for PH. PH was found in 137 of 386 patients, an incidence of 27% at 1 year and 48% at 2 years [19]. Interestingly, of the 93 patients that had clinically appreciable PH, nearly half had recurrent or metastatic disease. Half of these patients had no symptom data available, so it is hard to comment on whether larger, more clinically significant PH tend to be more symptomatic. On multivariable analysis, the authors found that female gender (HR 2.2), BMI (HR 1.1), and preoperative albumin (HR 0.4) were associated with the development of PH, similar to other series [30, 31]. They did not, unfortunately, comment on the symptomatic rate overall.
The repair of parastomal hernia is challenging and morbid, with a disappointing success rate. Given this, surgical repair should be avoided unless a patient is symptomatic and has failed conservative measures. These include belts and appliances that keep the hernia reduced. Surgical options include repairing the fascial defect versus relocating the stoma either primarily or with a biologic or synthetic mesh.
Primary repair has quite uniformly been shown to have a high failure rate. In a series of 94 patients with parastomal hernia, Rubin et al. found that recurrence rates were 76% for fascial repair, 33% with stomal relocation, and 50% with synthetic mesh, although this latter group was comprised of only seven patients [32]. Median time to recurrence was 29 months in the stoma relocation group, compared to only 13.5 and 11 in the other two, but there was no difference in reoperation rate, which was exceedingly high at 40%. Additionally, postoperative complication rates overall were over 60%. The authors concluded that even when primary fascial repair is abandoned for its dismal results, a durable surgical repair is rare, morbidity is high, and thus repair should be avoided if possible.
Mesh repair has shown better results regardless of approach. There are four possible techniques: onlay, in which the mesh is placed on top of the anterior fascia; inlay, in which mesh is sutured only to the sides of the fascial defect; sublay, in which mesh is placed between the rectus and posterior sheath; and an intraperitoneal onlay, in which mesh is affixed to the peritoneum. The latter can be done in an open or minimally invasive approach and often employs a biologic or two-layer mesh where the nonabsorbable side is affixed to the abdominal wall and an absorbable side is in contact with the abdominal contents in order to minimize mesh complications. Whereas the onlay and inlay techniques have not gained wide use due to the need for extensive dissection and high failure rate, respectively, the sublay technique has performed well [33].
Prophylactic placement of mesh at time of stomal creation is an option. To date, seven randomized trials have been performed in the general surgery arena examining this. Two used partially absorbable synthetic mesh in the setting of end colostomy. In one, at 1-year follow-up, parastomal hernia was present in 1 of 27 in the treatment arm compared to 13 in the control arm [34]. In the other, again with 27 patients in each arm but with a longer follow-up of 29 months, 14.8% in the treatment arm versus 40.7% in the control arm developed a hernia clinically and 22.2% versus 44.4% radiographically, respectively [34]. Another trial used biologic mesh for a loop stoma; at a short follow-up of 6 months, there were no hernias in the treatment group and 33.3% in the control group [35]. Although detailed complication data was not included in most of these, no mesh infections were noted in any trial.
Several studies have examined laparoscopic placement of parastomal mesh in an onlay (intraperitoneal) approach at time of ostomy creation. Lopez-Cano et al. randomized 36 patients at time of laparoscopic abdominoperineal resection and colostomy using synthetic mesh and at 1 year found parastomal hernia in 50% of the mesh group and 93.8% of patients in the control group [36]. Vierimaa et al. found, in a similar trial, no difference in the rate of radiographic PH at 1 year, but a decrease in clinically appreciable PH in the mesh (14.3%) versus control (32.3%) groups [37].
A recent meta-analysis of prophylactic mesh placement trials at time of end colostomy concluded that from six trials (156 mesh patients, 153 control patients) using either sublay or intraperitoneal, open or lap approaches, placement of mesh decreased the incidence of parastomal hernia by almost half (24.4% vs. 50.3%, respectively), (RR 0.65, 95% CI 0.33–1.30), with no difference seen between approach and no improvement in “overall stomal morbidity,” although reoperation rate was lower [34]. The shortcomings of this meta-analysis include the wide heterogeneity among these trials as well as a lack of detail regarding their criteria for stomal morbidity. From an economic perspective, once patients with a life expectancy of less than 1 year were excluded, prophylactic mesh was cost-effective.
To date there are no published randomized trials of prophylactic mesh placement at time of ileal conduit. In the only series published, Styrke et al. describe their experience with 114 consecutive patients undergoing synthetic mesh placement with a sublay approach at time of IC, with follow-up data for 58 patients [38]. They found a clinical parastomal hernia rate of 14% at 32 months and no mesh complications. Randomized controlled trials are needed in this area to confirm what the general surgery data for colostomy suggests, which is that mesh placement at time of IC creation reduces the rate of PH formation and is well tolerated.
Stomal Stenosis in the Incontinent Stoma
The incidence of stomal stenosis ranges from 2.8% to 19% in patients with ileal conduits and 10–20% in patients with colon conduits, with lower rates seen in two large contemporary series (8.5% and 2.8%, respectively) [4, 18]. While this incidence rate is lower than that seen in continent stomas, it still represents a significant problem and one that can be challenging to manage. By impairing drainage, stenosis can lead to hydronephrosis and stasis, increasing the risk for stones, renal compromise, and infection. An easily avoidable cause of stomal stenosis is stomal retraction, which can lead to difficulty with appliance placement and subsequent skin irritation, hyperkeratosis, and, eventually, stenosis.
A too small fascial opening can also lead to stenosis, both distally where ischemia may lead to stenosis in this distal segment, as well as ongoing pressure at the level of the fascia with resultant stenosis over time.
Loop stomas have a lower incidence of stenosis than end stomas. Care should be taken at time of stoma creation to bring the mucosa at least 2 inches above the skin. Again, we advocate for a fascial opening that just barely allows two fingers, which should be sufficient to avoid ischemia while minimizing the risk of hernia. In obese patients, a thick and short mesentery is often present. Although an end stoma is often advocated in the literature in this situation, we favor a loop stoma in these patients for easier management of a bulky mesentery. While ligation of the distal mesentery has been shown to be safe in animals, this does risk devascularization. Defatting the creeping fat off of the bowel itself can be helpful.
Another way to prevent retraction is to place anchoring sutures to the anterior rectus sheath, but care must be taken that there is no tension from the intraperitoneal side and account for postoperative distention in order to avoid bowel ischemia, which will be a setup for stomal retraction. We prefer placement of four circumferentially placed absorbable sutures. Proper eversion of the distal mucosa to form a nice, conical “rosebud”-type orientation is also important to avoid retraction and promote appliance fit. The placement of everting sutures can accomplish this.
Stenosis of the conduit has a higher incidence among obese patients; obesity likely contributes in several ways such as a thick abdominal wall and a shorter and thickened mesentery. These then contribute to poorly fitting devices as well as stomal retraction. The incidence of stomal retraction in ileal conduit is reported to range from 9Stomal Complications to 15% [14]. Panniculectomy at time of revision has been reported in this scenario. In a case series of four female patients, Katkoori et al. describe the procedure, which does not involve any intraperitoneal dissection, with good short-term outcomes: operative time of 2 h, LOS of 3 days, and no recurrences at 2-year follow-up [39].
Taneja et al. advocate maturing the stoma prior to transposing it to the skin to minimize stomal retraction [40]. This is done using an end stoma, defatting the distal mesentery and circumferentially everting the mucosa prior to bringing it up into the skin.
Another cause of stenosis is hyperkeratosis. This is a consequence of ongoing irritation to the surrounding peristomal skin, typically caused by a poorly fitting appliance, and is characterized by progressive hardening and proliferation of the skin and eventual stenosis of the stoma (see Fig. 9.2). Bacterial or fungal infections can also compound the problem. Hyperkeratosis can be managed by placing a catheter into the stoma to avoid urinary contact along with light therapy or topical vitamin C or 0.25% acetic acid washes [27]. Overall, early management with a combination of minimizing urine contact and skin treatments can reverse the problem, although if stenosis is significant, excision of the hyperkeratotic skin is an option. Again, perioperative consultation with an enterostomal therapist has been shown to decrease the development of stomal complications including chronic skin disorders [10].
Fig. 9.2
A severe stomal stenosis with surrounding hyperkeratotic skin
In a series of over 300 patients where stenosis was seen in 8.5%, 5% of patients required surgical management [18]. Surgical management can be divided into two groups, that at the skin level and that involving intraperitoneal mobilization, depending on the severity and location of stenosis. As such, a loopogram can aid in preoperative planning. At the skin level, the simplest option involves making a circumferential incision around the stoma and mobilizing the conduit to the level of the fascia, excising the stenotic ring and hyperkeratotic skin, and reanastomosing to the skin. A local V flap can alternatively be rotated into the defect. However, these repairs do not address stomal retraction and may even lead to it [27]. Another option, again using a circumscribing incision at the stoma, is to continue mobilization intraperitoneally from this approach, to allow sufficient conduit length to be advanced. If intraperitoneal mobilization cannot be achieved through the existing fascial opening, it can be extended, although this may increase the risk of parastomal hernia development. Rarely, full laparotomy is required to gain sufficient mobilization of the conduit. Of note, one must be careful not to compromise the mesentery during the superficial mobilization regardless of approach.
Stomal Complications in Continent Cutaneous Diversion
While much less commonly performed than the conduit, the continent cutaneous diversion is an important option for patients desiring continent diversion in whom an orthotopic substitution is contraindicated or in whom the potential risks of an orthotopic diversion such as incontinence, retention, or fistula are high. Although up to 80% of men and 65% of women have been shown to be suitable candidates for continent diversion after cystectomy [41], it is well established that the rates of continent diversion performed in this setting are far less. A review of over 5,000 cystectomies in the United States found a continent diversion rate of 14.3% [42]. A subsequent review of the same database found only an 8.3% incidence of continent diversions [43]. Of these only a small fraction was CCD. The most likely explanation of why CCD is employed relatively infrequently even at large volume centers is that it is more complex to construct and is perceived to carry a higher rate of complications than other diversions. In fact, a meta-analysis of several series using a host of different types of CCD has shown continence rates overall to be excellent, ranging from 83% to 100% [44]. Several papers in addition to this have shown acceptable complication rates, similar to those seen with a conduit. However, multiple quality of life studies have not shown a significant difference between diversion types overall, thus weighing against the benefit of CCD [45]. One study comparing IC and CCD found that 41% versus 71%, respectively, reported being satisfied with their diversion [46].
Overall the CCD is an important option for a certain subset of patients with reasonable overall complication rates, and as such, it is an important part of the urologist’s armamentarium. It does carry a unique set of complications, namely, those involving the efferent channel, that is, the stoma and the catheterizable channel, which will be the focus of discussion here. There are a multitude of different forms of CCD; we will focus on complications as they arise in the most commonly performed types.
Preventing Complications
Appropriate patient selection is extremely important prior to undertaking a CCD. The indications for CCD are any patient who desires a continent diversion but for whom an orthotopic diversion is contraindicated or not suitable. This includes patients with a positive urethral margin, prior pelvic radiation, locally extensive disease, extensive urethral stricture, renal insufficiency (defined at Cr > 1.8 mg/dl or GFR < 40 ml/min/1.73 m2), hepatic insufficiency, chronic inflammatory bowel disease, and neurologic impairment of dexterity [47]. In addition to these, for some patients, the functional outcomes of an orthotopic diversion may not be desirous; this usually pertains to continence outcomes. In females, where up to 40% will be in retention requiring intermittent catheterization and up to 30% will be significantly incontinent, a CCD may be more attractive than an orthotopic diversion [48]. Men with existing stress incontinence may similarly fare better with a CCD if they desire a continent diversion.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
